Mohsen Guizani

Prima.cpp: Fast 30-70B LLM Inference on Heterogeneous and Low-Resource Home Clusters

Uncertainty Quantification for Regression Using Proper Scoring Rules

Multidimensional Uncertainty Quantification via Optimal Transport

Client-Aware Multimodal Distillation with Adaptive Aggregation for Robust Federated Learning in Noisy and Adversarial Environments

FRL-SAGE: Stackelberg Game-Theoretic Defense Against Adaptive Adversaries in Federated Reinforcement Learning

Probabilistic Conformal Prediction with Approximate Conditional Validity

Rectifying Conformity Scores for Better Conditional Coverage

TPI-LLM: Serving 70B-scale LLMs Efficiently on Low-resource Edge Devices